[tor-bugs] #21419 [Metrics/Metrics website]: avoid spurious downturns and 'spamming' of top-10-censorship list

[Tor Bug Tracker & Wiki]

#21419: avoid spurious downturns and 'spamming' of top-10-censorship list
-------------------------------------+------------------------------
 Reporter:  iwakeh                   |          Owner:  metrics-team
     Type:  defect                   |         Status:  needs_review
 Priority:  Medium                   |      Milestone:
Component:  Metrics/Metrics website  |        Version:
 Severity:  Normal                   |     Resolution:
 Keywords:                           |  Actual Points:
Parent ID:                           |         Points:
 Reviewer:                           |        Sponsor:
-------------------------------------+------------------------------
Description changed by iwakeh:

Old description:

> [https://metrics.torproject.org/userstats-censorship-
> events.html?start=2016-01-01&end=2016-12-31 top 10 list for 2016]
> [[Image(top10censorship-2016-marked.png)]]
> The marked countries have a very low client count and the predicted lower
> value is zero.
> Starting from the top, Christmas Island has a median user count of one
> in 2016 with the third quartile being two.  Also, Netherlands Antilles
> have a median of two connections, Kiribati, British Indian Ocean
> Territory, Norfolk Island, and Tokelau each a 2016 user count median of
> one.
>
> Similarly, the downturns in country specific censorship events, for
> example [https://metrics.torproject.org/userstats-relay-
> country.html?start=2016-01-01&end=2016-12-31&country=an&events=on
> Netherlands Antilles in 2016] [[Image( userstats-relay-country-
> an-2016-03-01-2016-09-30-on.png)]]
>
> I attached a simple path.

New description:

 [https://metrics.torproject.org/userstats-censorship-
 events.html?start=2016-01-01&end=2016-12-31 top 10 list for 2016]
 [[Image(top10censorship-2016-marked.png)]]
 The marked countries have a very low client count and the predicted lower
 value is zero.
 Starting from the top, Christmas Island has a median user count of one
 in 2016 with the third quartile being two.  Also, Netherlands Antilles
 have a median of two connections, Kiribati, British Indian Ocean
 Territory, Norfolk Island, and Tokelau each a 2016 user count median of
 one.

 Similarly, the downturns in country specific censorship events, for
 example [https://metrics.torproject.org/userstats-relay-
 country.html?start=2016-01-01&end=2016-12-31&country=an&events=on
 Netherlands Antilles in 2016] [[Image( userstats-relay-country-
 an-2016-03-01-2016-09-30-on.png)]]

 I attached a simple patch.

--

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/21419#comment:4>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online